How to Read Your CGM Data

A continuous glucose monitor produces thousands of readings a day, which can feel like a wall of numbers. The good news is that most modern reports are designed to make that data legible at a glance. This guide walks through the parts you'll actually look at — the summary report, the median curve and its spread, the overnight window, post-meal rises, and the Time in Range bars — and the kinds of patterns worth noticing.

Start with the standardized report

Rather than scrolling through raw traces day by day, most clinicians and devices summarize CGM data on a single standardized page called the Ambulatory Glucose Profile (AGP). It condenses a week or two of readings into a few consistent elements so the same layout is recognizable whichever sensor produced it. If you are new to the whole topic, the CGM guide covers the fundamentals of how these devices sample glucose in the first place, and how CGM works explains what the sensor is actually measuring beneath the skin.

A typical report opens with a handful of headline figures — average glucose, an estimated indicator sometimes shown as a Glucose Management Indicator, the percentage of time your glucose stayed in a target band, and a measure of variability. Read these as a starting orientation, not a verdict.

The median line and the spread

The centrepiece of the AGP is a single 24-hour chart. Instead of plotting each day separately, it overlays many days on top of one another and draws a median line — the middle value for each time of day — with shaded bands around it showing how widely readings scatter above and below.

Two things are worth reading here. First, the height of the median line tells you the typical glucose level at a given hour. Second, the width of the shaded band tells you how variable glucose is at that hour. A narrow band means your glucose behaves consistently at that time; a wide band means it swings a lot. A tall, wide bulge in the late morning, for example, might point to a variable response to breakfast that a single average would never reveal.

The overnight window

The stretch from roughly midnight to waking is often the most informative and the most overlooked. Because you are usually not eating or exercising, overnight glucose reflects the body's baseline handling of sugar with fewer confounders. A median line that sits flat and steady overnight is generally reassuring; a line that drifts, dips, or climbs may prompt questions worth raising with a clinician. Interpretation still depends on your individual circumstances, so treat overnight patterns as context rather than conclusions.

Post-meal rises

After eating, glucose normally rises and then settles back down over a couple of hours. On the AGP, meals often appear as recurring bulges in the median line at the times you typically eat. What clinicians tend to look at is the size of the rise and how long it takes to return toward baseline. Sharper, more prolonged rises can be a topic to explore — though a single high reading after an unusual meal is rarely meaningful on its own. Because CGM measures glucose in fluid just beneath the skin rather than in blood directly, there is also a short lag; CGM accuracy and CGM versus fingerstick both go deeper on why a sensor value and a finger-prick can differ in the moment.

Time in Range bars

Most reports include a stacked bar showing the proportion of readings that fell below, within, and above target. This Time in Range summary is popular because it captures something an average cannot: whether steady numbers hide frequent highs and lows that roughly cancel out. Reading the bar is straightforward — more of the bar in the target band is generally the goal — but the specific targets that are appropriate for you are a clinical decision, not a universal rule.

Patterns worth looking for

Once the elements make sense, reading a report becomes a matter of scanning for recurring shapes rather than reacting to any single point:

• Consistency vs. chaos — do the bands stay narrow, or do they fan out widely at certain hours?
• Time-of-day themes — do highs cluster after particular meals, or lows in the small hours?
• Repetition — a pattern that shows up on most days carries more weight than a one-off.

The aim of reading your own data is not self-diagnosis but better questions. Patterns you spot are best brought to a qualified clinician, who can interpret them alongside your history and any standard testing. Endobits is used as clinical decision support to help clinicians review this kind of data — it does not diagnose or treat on its own.

Sources

American Diabetes Association, diabetes.org, on glucose targets and Time in Range. Centers for Disease Control and Prevention, cdc.gov/diabetes, on monitoring blood glucose. Battelino T. et al., "Clinical Targets for Continuous Glucose Monitoring Data Interpretation," Diabetes Care (2019), on standardized CGM metrics and the AGP.

Related: The Ambulatory Glucose Profile explained · What is Time in Range? · HbA1c vs. CGM